Autopsy-proven Determinants of Death in HIV-infected Patients Treated for Pulmonary Tuberculosis in Sao Paulo, Brazil

https://doi.org/10.1078/0344-0338-00264Get rights and content

Summary

The aim of this work is to describe and compare pulmonary pathology and proven causes of death in HIV-infected and non-HIV patients treated for tuberculosis, to identify the presence and extension of the lesions, and to suggest appropriate interventions based on the results. Of 246 adult patients (aged > 15) autopsied with tuberculosis and tested for HIV infection at Hospital das Clinicas, School of Medicine, São Paulo University, from January 1994 to December 1996, 100 HIV-infected patients were matched with 44 non-HIV-infected patients. Demographic determinants influencing patients' death were as follows:

1. Age and sex were not found to be important for the histological outcome, but do seem to correlate with HIV infection. Older patients with tuberculosis are less likely to be HIV-infected;

2. Previous tuberculosis and its treatment had no influence on the course of secondary or reinfection tuberculosis;

3. The efficiency of the diagnostic criteria used at the time of death was very low (59.2%);

4. Tuberculosis was more frequently investigated in HIV-infected than in non-HIV-infected patients;

5. Only 79 (56%) of the patients received first line agents for treatment;

6. Patients presented with advanced disease;

7. Their mortality is high, and death occurs early.

Morphological determinants with influence on patient's death were related to differences in the spectrum of tuberculosis presentation and time of treatment. Non-caseating generalized multibacillary tuberculosis was likely to be the primary cause of death in HIV-infected patients who died during therapy, whereas deaths of patients occurring after the second course of treatment because of recurrence or incomplete treatment were increased for other manifestations of HIV disease (pyogenic pneumonia, Pneumocystis carinii pneumonia, cerebral toxoplasmosis, wasting syndrome). In these patients, dimorphic tuberculosis, an intermediate reactive spectrum form of presentation, was the predominant histological finding. In the opposite spectrum, paucibacillary tuberculosis, a reactive form of tuberculosis presentation, was equally regarded in non-HIV-infected patients as the primary cause of death after four months of therapy. In the same spectrum, non-HIV-infected patients with tuberculosis, who completed all or most of their treatment, died of associated diseases after therapy (alcoholism, cancer, diabetes mellitus). Autopsy-proven determinants of death were associated with HIV status, the spectrum of tuberculosis presentation, and time of treatment. Early treatment for tuberculosis and associated diseases can improve survival and the quality of life even of highly immunosuppressed patients.

References (33)

  • A.M. Dannenberg et al.

    Liquefaction of the caseous foci in tuberculosis

    Am Rev Respir Dis

    (1976)
  • A.E. Greenberg et al.

    Autopsy = proven causes of death in HIV-infected patients treated for tuberculosis in Abidjan, Côte d'Ivoire

    AIDS

    (1995)
  • B.F. Hansen

    Pathology of the heart in AIDS

    APMIS

    (1992)
  • A.R. Hill et al.

    Isseminated tuberculosis in the acquired immunodeficiency syndrome era

    Am Rev Respir Dis

    (1991)
  • S. Kassin et al.

    Twoyear follow-up of persons with HIV-1 and HIV-2-associated pulmonary tuberculosis treated with short-course chemotherapy in West Africa

    AIDS

    (1995)
  • D.N. Lanjewar et al.

    Major differences in the spectrum of gastrointestinal infections associated with AIDS in India versus the West: an autopsy study

    Clin Infect Dis

    (1996)
  • Cited by (25)

    • HIV-1 and the Mycobacterium tuberculosis granuloma: A systematic review and meta-analysis

      2016, Tuberculosis
      Citation Excerpt :

      Four studies did not differentiate results between Mtb and other mycobacteria [68–71]. One study may have been biased by persons having co-morbidities [72]. Two studies [31,33] reanalyzed, at least in part, some results from persons in previous studies [73,74].

    • Clinical Features and Management of HIV/AIDS

      2013, Manson's Tropical Diseases: Twenty-Third Edition
    • Factors associated with mortality and default among patients with tuberculosis attending a teaching hospital clinic in Accra, Ghana

      2011, Transactions of the Royal Society of Tropical Medicine and Hygiene
      Citation Excerpt :

      The use of treatment supporters to provide social support and facilitate directly observed ingestion of drugs was associated with a high treatment success rate in our study as well as other prior studies.17,18 Our findings are consistent with those of large randomized studies that have shown that DOTS by community health care workers or persons outside the family selected by the patient results in improvement of treatment outcomes.31–33 Our findings show that the use of treatment supporters has a good impact in our settings and all efforts should be made to identify treatment supporters for all patients in our treatment programme including the HIV co-infected.

    View all citing articles on Scopus

    This project was submitted and approved by the Department of Pathology, School of Medicine, University of São Paulo.

    This study was supported by the following Brazilian agencies: the National Council for Scientific and Technological Development (CNPq); The Foundation for the Support of Research of the State of São Paulo (FAPESP); and the Laboratories for Medical Research (LIMs), Clinicas Hospital, School of Medicine, University os São Paulo.

    View full text